Method and apparatus for predicting braking system friction

Abstract

A brake system control method determines vehicle operating conditions, compares the conditions to an allowable range, and uses a neural network to predict an expected coefficient of friction when the conditions are within the range. When the conditions fall outside of the range, the method determines an amount of required braking force using a constant coefficient of friction, and calculates the required braking force using the expected coefficient of friction when the conditions are within the range. The vehicle operating conditions include a vehicle speed, brake pressure, modeled brake rotor temperature, and apply state. The expected coefficient is multiplied by a constant or a calculated correction factor. A vehicle has an engine, transmission, and braking system, with a controller and an algorithm for predicting a coefficient of friction for two brake rotors, calculating a hydraulic brake pressure, and for applying the braking system using the hydraulic brake pressure.

1. A method for determining a required braking force in a vehicle having a brake rotor, the method comprising:
 determining values for a set of vehicle operating conditions, including determining a bulk temperature of the brake rotor; 
 comparing said set of vehicle operating conditions to an allowable range; 
 predicting, in real time using a neural network, an expected coefficient of friction at a friction interface of the brake rotor and using the set of vehicle operating conditions, said expected coefficient of friction corresponding to said set of vehicle operating conditions when said set of vehicle operating conditions are within said allowable range; 
 determining an amount of the required braking force using a constant coefficient of friction value when said set of vehicle operating conditions are not within said allowable range; and 
 calculating the required braking force using said expected coefficient of friction when said set of vehicle operating conditions are within said allowable range. 
 
     
     
       2. The method of  claim 1 , wherein said predicting an expected coefficient of friction includes processing each of said set of vehicle operating conditions through a different input node of an input layer of the neural network. 
     
     
       3. The method of  claim 1 , further comprising modeling a thermal profile of the brake rotor as the bulk temperature. 
     
     
       4. The method of  claim 3 , wherein said set of vehicle operating conditions further includes a speed of the vehicle, a hydraulic braking pressure, and a braking system apply state. 
     
     
       5. The method of  claim 1 , further comprising:
 multiplying said expected coefficient of friction by an error correction factor prior to determining the required braking force. 
 
     
     
       6. The method of  claim 5 , wherein said calculating said error correction factor includes:
 calculating an average value for said expected coefficient of friction over a predetermined sample size; 
 calculating an average value for an actual coefficient of friction over said predetermined sample size, said actual coefficient of friction being determined at least partially from a recorded deceleration response of the vehicle; 
 multiplying said apparent coefficient of friction by a first error correction value when a difference between said average values for said apparent and said actual coefficients of friction is less than or equal to a threshold value; and 
 multiplying said apparent coefficient of friction by a second error correction value when said difference is greater than or equal to said threshold value. 
 
     
     
       7. A method for optimizing a blended braking event of a vehicle having a controller configured to selectively apply an electronic braking torque via a motor and a hydraulic braking force via a master cylinder during the blended braking event, the method comprising:
 providing the vehicle with a neural network having an input layer, at least one hidden layer, and an output layer; 
 determining values for a set of vehicle operating conditions, including at least a bulk temperature of a brake rotor of the vehicle; 
 feeding each condition in said set of vehicle operating conditions forward into a corresponding input node of said input layer of said neural network; 
 using said neural network for predicting an expected coefficient of friction at a friction interface of the brake rotor, in real time, wherein said expected coefficient of friction corresponds to said set of vehicle operating conditions when said set of vehicle operating conditions are within an allowable range; 
 determining an amount of a required braking force using a constant coefficient of friction value when said set of vehicle operating conditions are not within said allowable range; 
 calculating said required braking force using said expected coefficient of friction when said set of vehicle conditions are within said allowable range; and 
 allocating the requested braking torque, via the controller, between the electronic braking torque and the hydraulic braking pressure to thereby execute the blended braking event. 
 
     
     
       8. The method of  claim 7 , wherein said determining said set of vehicle operating conditions includes modeling a thermal profile of said at least one brake rotor as the bulk temperature, and wherein said set of vehicle operating conditions includes the bulk temperature of the brake rotor as determined from said modeling. 
     
     
       9. The method of  claim 8 , wherein said determining said set of vehicle operating conditions further includes at least one of measuring a speed of the vehicle, detecting a brake apply pressure, and a determining an apply state of the braking system. 
     
     
       10. The method of  claim 7 , including configuring said hidden layer with a plurality of tan-sigmoid neurons, and feeding a set of output values from said input set into said plurality of tan-sigmoid neurons. 
     
     
       11. A vehicle comprising:
 a braking system having a set of front brake rotors, a set of rear brake rotors, a hydraulic braking mechanism, and an electronic braking mechanism, wherein the hydraulic braking mechanism and the electronic braking mechanism may be applied during a blended braking event; and 
 a controller having a control algorithm and a neural network for controlling said braking system in response to a requested braking torque , wherein the controller is configured to selectively apply the electronic braking torque as an opposing torque to at least one of the output member and the front or rear brake rotors, and to apply a hydraulic braking force to at least one set of the front and rear brake rotors via the hydraulic braking mechanism, during the blended braking event; 
 wherein said neural network is adapted for receiving a set of vehicle operating conditions including at least a bulk temperature of the brake rotors, and using said set of vehicle operating conditions for predicting, in real time via the neural network, an expected coefficient of friction for each of said sets of front and said rear brake rotors; and 
 wherein said controller is configured for allocating the requested braking torque between the electronic braking torque and the hydraulic braking force to thereby execute the blended braking event. 
 
     
     
       12. The vehicle of  claim 11 , wherein said controller is operable for applying said hydraulic apply mechanism according to said hydraulic apply pressure that is calculated by said control algorithm. 
     
     
       13. The vehicle of  claim 12 , further comprising:
 at least one sensor operable for directly detecting said hydraulic apply pressure, a speed of the vehicle, and an apply state of the braking system; and 
 a brake thermal model configured for modeling a thermal response of each of said front and rear brake rotors to thereby determine the bulk temperatures; 
 wherein said set of vehicle operating conditions includes said hydraulic apply pressure, said speed, said apply state, and said thermal response. 
 
     
     
       14. The vehicle of  claim 11 , wherein the controller includes an error correction model for comparing an average value of said expected coefficients of friction to a corresponding average value of a calculated coefficient of friction to determine an error value therebetween, wherein the controller is configured for adjusting said expected coefficients of friction proportionately to said error when said error exceeds a predetermined confidence level. 
     
     
       15. The method of  claim 1 , wherein the vehicle includes a controller, a motor, and a master cylinder, the method further comprising:
 automatically allocating the requested braking torque, via the controller, between an electronic braking torque from the motor and a hydraulic braking force from the master cylinder to thereby execute a blended braking event.